Convergent Ditopic Receptors Enhance Anion Binding upon Alkali

Org. Lett. , Article ASAP. DOI: 10.1021/acs.orglett.8b03778. Publication Date (Web): January 14, 2019. Copyright © 2019 American Chemical Society. *E...
0 downloads 0 Views 872KB Size
Letter Cite This: Org. Lett. 2019, 21, 652−655

pubs.acs.org/OrgLett

Convergent Ditopic Receptors Enhance Anion Binding upon Alkali Metal Complexation for Catalyzing the Ritter Reaction Kang Kang,† Jessica A. Lohrman,‡ Sangaraiah Nagarajan,† Lixi Chen,† Pengchi Deng,† Xin Shen,† Kuirong Fu,† Wen Feng,*,† Darren W. Johnson,*,‡ and Lihua Yuan*,† †

Downloaded via SWINBURNE UNIV OF TECHNOLOGY on February 3, 2019 at 21:52:05 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

Key Laboratory for Radiation Physics and Technology of Ministry of Education, Institute of Nuclear Science and Technology, College of Chemistry, Analytical & Testing Center, Sichuan University, Chengdu 610064, China ‡ Department of Chemistry & Biochemistry and the Materials Science Institute, University of Oregon, Eugene, Oregon 97403-1253, United States S Supporting Information *

ABSTRACT: A supramolecular approach to catalyzing the Ritter reaction by utilizing enhanced anion-binding affinity in the presence of alkali metal cations was developed with ditopic hydrogen-bonded amide macrocycles. With prebound cations in the macrocycle, particularly Li+ ion, their metal complexes exhibit greatly enhanced catalytic activities. The catalysis is switchable by removal or addition of the bound cation. The method described in this work may be generalized for use in other anion-triggered organic reactions involving heteroditopic receptors capable of ion pairing.

T

he past decade has witnessed a growing wealth of ditopic receptors that are capable of cooperatively binding pairs of cations and anions.1 Compared to a monotopic molecular receptor2 that only recognizes a single ion, these receptors show enhanced binding affinity and selectivity, leading to interesting applications which include salt solubilization,3 extraction,4 ionsensing, and transmembrane ion transport.5 So far, ion-pair binding has been explored in a variety of scaffolds including crown ethers,6 calix[4]pyrroles,7 and urea derivatives.8 In particular, hosts that utilize convergent binding sites have shown a significant increase in anion affinity and selectivity in the presence of alkali metal cations.9 While direct complexation of inorganic salts requires both ionic species to be bound to two specific sites, a cation-enhanced anion binding process involves addition of a metal cation first, followed by introduction of an external anionic guest to the core of the receptor system. This would be useful in catalyzing a reaction when the mechanism calls for anion sequestration. Despite many reports on research into enhanced anion binding in the presence of alkali metal ions with ditopic receptors,10 none of these have been developed for catalytic reactions. Recently, anion-binding receptors have attracted particular interest in supramolecular catalysis.11 Catalysts that require anion sequestration generally employ hydrogen-bonding interactions based on strongly polarized N−H,12 O−H,13 aryl C−H14 and halogen bonds.15 Our previous work with hydrogen-bonded (H-bonded) amide macrocycles16 (Figure 1) showed that convergent heteroditopic cyclo[6]aramides17 were able to coordinate organic ion pairs. This is ascribed to a © 2019 American Chemical Society

Figure 1. Molecular structures of hydrogen-bonded receptors with protons labeled and their cartoon representation and alkali metal ions/ anions involved in the catalytic reaction.

convergent binding site composed of two amide N−Hs and an aryl C−H serving as H-bond donors for anions and four amide carbonyl oxygens for cations. However, use of H-bonded aromatic amide macrocycles18 for binding anionic guests to achieve catalytic reactions has reamined unknown. Herein, we present a supramolecular strategy for catalysis of the Ritter reaction by using convergent H-bonded amide macrocycles as receptors with prebound alkali metal to enhance anion affinity. Importantly, the catalytic process was found to be switchable in an on-and-off manner triggered by external addition of inorganic salts. To this end, two new ditopic receptors (1a and 1b), bearing −NO2 and −OMe groups with contrasting electronic properties, were synthesized (Figures Received: November 26, 2018 Published: January 14, 2019 652

DOI: 10.1021/acs.orglett.8b03778 Org. Lett. 2019, 21, 652−655

Letter

Organic Letters

Table 1. Association Constants Ka (M−1) of Complexes in CDCl3/CD3CN (1:1, v/v) at 298 K

S1−S9). The anion-binding effect with the presence of alkali metal ions was probed first, followed by employing the cation complex of the heteroditopic receptors as anion-binding catalysts for promoting the Ritter reaction. Anion affinity for halide guests (Cl−, Br−, and I−) was initially determined (Figure 2) in the absence of alkali metal cations via 1 H NMR spectroscopy studies in a 1:1 solution of CDCl3 and CD3CN using tetrabutylammonium (TBA) salts, since TBA is too large to fit in the ditopic receptor binding pocket (Figure S11). Downfield shifting of amide NH (Hj) and aromatic ArH resonances (Hk) in receptor 1a was observed only for Cl−, while H-bond donor resonances remained unmoved upon addition of Br− and I−, indicating receptor selectivity toward Cl−. As expected, receptor 1b containing the electron-donating group (OMe) shows a relatively smaller chemical shift upon anion addition owing to the decreased acidity of amide protons (Figure S12). The receptors were also assessed for alkali metal ions Li+, Na+, and K+ in the form of their perchlorate (ClO4−) and hexafluorophosphate (PF6−) salts (Figures S10 and S15− S17). The response to either anions or cations indicates sitespecific binding by these convergent heteroditopic receptors. Subsequently, anion-binding properties for halide guests were evaluated (Figure 2) in the presence of alkali metal cations. Addition of Br− to a solution of 1a·Li+ in a 1:1 solution of CDCl3 and CD3CN leads to 0.66 and 0.61 ppm downfield shifts of Hj and Hk (Figure 2). Tests of 1b·Li+ with Br− followed the same trend (Figure S14). Typically, shifting observed in the presence of Li+ increases by over 6-fold compared to the shifting in the absence of these metal ions (Figure S13). This indicates that prebound metal cations can enhance the complexation of anions by these ditopic receptors. Recently, the Johnson and Haley groups also revealed a similar phenomenon with a phosphine oxide based ditopic receptor.19

a b

receptor

Cl−

Br−

I−

1a 1a·Li+ 1a·Na+ 1a·K+

110 2300 4000 4900

a 1100,b 55000b 4200 3800

a 9800 a a

K a < 10 M −1 and could not be accurately determined. Stoichiometry was 2:1.

Ka for all of the halide anions in comparison to the metal-free system. The enhanced affinity afforded by the presence of Li+ allowed for an accurate determination of the Ka of Cl− (Ka = 2300 M−1), Br− (K1 = 1100 M−1, K2 = 55000 M−1), and I− (Ka = 9800 M−1), which are otherwise not observed in the absence of Li+ salt. A 1:1 binding stoichiometry for Cl− and Br− in the presence of Na+ and K+ was derived from the 1H NMR titration curves for 1a and 1b (Figures S24−S53). Furthermore, job plot experiments suggested a 2:1 stoichiometry between the complex of 1a·Li+ and Br− (Figure S32). While significant increases in Ka for Cl− and Br− persisted in the presence of Na+ and K+ cations, I− remained unbound. The inability to effectively bind I− is attributed to the confined cavity restricting the size of the halide guest after the coordination of larger metal ions. With no metal ion present, 1a was unable to engulf I−; however, after the addition of smaller-sized Li+, I− was strongly bound with a Ka as high as 9800 M−1. Therefore, I− could only be accommodated alongside a smaller cation (Li+) where sufficient space was still available for the larger anionic guest. These findings were echoed by a similar ion-size trend that was observed for 1b (Table S2).The enhanced anion affinity is attributed to cooperative action from hydrogen bonding of amide NH with the anion and electrostatic attraction from the metal cation within the cavity, the later of which is corroborated by the loss of binding of Br− and I− with the receptor 1a alone, and the inability of 1b to bind any halides in the absence of alkali metals. The affinity increases between 1a and 1a·cation complexes are further supported by the results from the Gibbs free energies (ΔΔG) (Table S2). The strong halide binding exhibited by macrocycles 1a and 1b in the presence of alkali metals makes them attractive as supramolecular catalysts. The catalytic potential of the hosts was investigated by using the Ritter reaction of bromodiphenylmethane (2a) and acetonitrile as a benchmark reaction. The receptors are expected to promote cleavage of the carbon− bromide bond, wherein the Ritter product, benzhydryl acetamide (2b), is formed. Reactions were run in the presence of stoichiometric amounts of catalyst (20 mol %) at room temperature for 5 days (Figures S58−S60). The resulting yields of benzhydryl acetamide are reported in Table 2. A control reaction composed solely of 2a in a 1:1 solution of CH2Cl2 and CH3CN was performed and failed to effectively afford Ritter product 2b in the absence of catalyst (Table 2, entry 1). The addition of metal-free 1a also failed to catalyze the reaction (Table 2, entry 2) due to its weak affinity for Br−. Subsequently, a series of experiments involving the alkali metal (Li+, Na+, and K+) complexes of 1a and 1b as the catalysts were carried out under the same conditions as the control experiments (Table 2, entries 3−12). In the presence of 1 equiv of alkali cations, the reaction proceeded to a significant extent as indicated by the yields of 42% (1a·Li+), 18% (1a·Na+), and 16% (1a·K+), with Li+ exhibiting the highest yield (Table 2,

Figure 2. Partial 1H NMR spectra (400 MHz, CDCl3/CD3CN, v/v = 1/1, 298 K) of receptor 1a and 1a·Li+ with 2 equiv of halide ions. (●) = Hj, (Δ) = Hk.

To quantify the anion-binding ability of receptors 1a and 1b in the absence or presence of alkali metal cations, 1H NMR spectroscopy titration experiments were performed in a 1:1 solution of CDCl3 and CD3CN and the downfield shifts of the amide resonances (Hj) were tracked after addition of TBA salts of the halides. The binding constants (Table 1 and Table S2) were determined by fitting to a 1:1 binding model for all experiments except for 1a·Li+ and 1b·Li+ binding with Br−, which was fit to a 2:1 binding model20 (Figures S18−S55). The anions affinities for 1a·Li+ were determined using lithium perchlorate as a Li+ source and yielded a noticeable increase in 653

DOI: 10.1021/acs.orglett.8b03778 Org. Lett. 2019, 21, 652−655

Letter

Organic Letters Table 2. Investigation of the Anion-Binding-Catalyzed Ritter Reaction with Various Potential Catalystsa

entry

catalyst

yieldc (%)

1 2 3 4 5 6 7 8 9 10 11 12

b 1a 1a·Li+ 1a·Na+ 1a·K+ 1a + CF3CO2H (20%)d 1a·(10 equiv of Li+) 1a·Li+ + CF3CO2H (20%)d 1b·Li+ 1b·(10 equiv of Li+) 1b·Na+ 1b·K+